Update README.md

README.md

@@ -1,20 +1,42 @@
 ---
-title: FastLLM 150M Demo
+title: FastLLM
 emoji: ⚡
-colorFrom: indigo
+colorFrom: blue
 colorTo: purple
 sdk: gradio
-sdk_version: 5.31.0
+sdk_version: "5.38.0"
+python_version: "3.10"
 app_file: app.py
 pinned: false
-license: apache-2.0
-short_description: Text generation demo for devoppro/FastLLM (150M)
 ---
 
-# FastLLM (150M) Demo
+# ⚡ FastLLM
 
-Interactive text-generation demo for [devoppro/FastLLM](https://huggingface.co/devoppro/FastLLM),
-a ~150M parameter decoder-only causal LM (GQA, SwiGLU, RMSNorm, RoPE) trained from scratch.
+Chat with FastLLM directly in your browser.
 
-Runs on ZeroGPU — the model loads once and `@spaces.GPU` requests a GPU
-slice only for the duration of each generation call.
+FastLLM is a ~150M parameter decoder-only causal language model created by **devoppro**.
+
+## Model
+
+- Parameters: ~150M
+- Architecture: ModernLLMForCausalLM
+- Hidden size: 768
+- Layers: 12
+- Attention: GQA
+- Query heads: 12
+- KV heads: 4
+- MLP: SwiGLU
+- Normalization: RMSNorm
+- Position encoding: RoPE
+- Context length: 2048
+- Format: Safetensors
+
+## Model
+
+https://huggingface.co/devoppro/FastLLM
+
+## Runtime
+
+This Space uses Hugging Face Transformers with the model's custom architecture.
+
+No AirLLM or flash-attn is required.